0 Fork 0

mirror of https://github.com/paperclipai/paperclip.git synced 2026-11-10 12:37:55 +01:00

Files

T

DottaandPaperclip 505e1a5582 feat(cases): add persistent workflows with same-task agent handoffs (#16959 )

## Thinking Path

> - Paperclip manages AI agents and their work.
> - An agent workflow needs persistent facts and a safe way to transfer
work.
> - Creating a task for each phase splits the objective, discussion,
workspace, and review policy.
> - A case can retain the facts while one primary task moves between
agents.
> - Transfers must survive process failure and preserve private access,
budgets, and completion gates.
> - This pull request adds durable case handoffs, native case tools, and
read-only case pages.
> - Prompts describe the workflow. Existing task controls decide when
work can run and finish.

## Linked Issues or Issue Description

**Problem or motivation**

Prompt-driven agent teams need a persistent record that can move through
triage, implementation, verification, and correction. Ordinary
reassignment does not provide a durable instruction, safe release
barrier, or recoverable delivery receipt for an agent handing off its
own running task.

**Proposed solution**

Extend standalone cases with one primary task, a descriptive phase,
retained access sources, and a handoff ledger. Apply assignments after
source execution and workspace settlement. Deliver each transfer through
the existing durable wake queue. Provide `create_case`, `read_case`,
`update_case`, and `handoff_case` in the native runner. Creation
requires kickoff and explicit workflow text. An explicit handoff to
`done` closes the case when normal task completion is accepted.

**Alternatives considered**

Do not add a workflow engine, filesystem queue, configured state graph,
or one task per phase. The workflow document combines transition prose
with an optional Mermaid `stateDiagram-v2` map. The server does not
execute the diagram. The Pipelines experiment removed on master stays
removed.

**Roadmap alignment**

This extends Agent Reviews and Approvals, Enforced Outcomes, Bounded Run
Recovery, and Activity Log attribution. Existing approval, review,
budget, privacy, workspace, and responsible-user rules remain
authoritative.

Related work: [ordinary handoff wake
recovery](https://github.com/paperclipai/paperclip/pull/15019). This
feature adds a separate persisted case-handoff receipt and release
barrier.

## What Changed

- Add case routing fields, retained access sources, durable handoff
history, delivery leases, and replay-safe migrations.
- Keep the same primary task across agent phases. Preserve each
returning agent's own session and the task's workspace.
- Fence outgoing mutations, wait for provider and workspace release, and
recover delivery without duplicate task comments or runs. Recheck
current assignment permissions as the original caller before applying a
delayed transfer. Scope a run Stop to that run’s own handoff intents, so
repeated old Stop requests preserve newer work. Cancelling terminal
intent without a source run also requires fenced current task ownership,
including native retries and preparation with an old checkout. For Cloud
callers, revalidate current stack membership with an uncached Cloud RPC
and derive owner elevation from the current managed setting.
- Keep case reads, documents, assets, Audit, and live events inside
retained task/project permissions.
- Add native case creation/read/update/handoff tools. Store required
workflow text and initial documents on the first task. Page oversized
case facts and referenced document bodies through the dedicated tool
with bounded UTF-8 JSON, current routing versions, snapshot conflicts,
and current source authorization on every page.
- Add read-only case lists and detail pages, task-style list controls,
compact dotted field tables, linked handoff tasks, and case filters in
Audit.
- Add native/legacy integration tests, migration/restart tests, browser
coverage, and bounded live receipt fixtures. Extend the trusted sandbox
condition to cover the new subscription fixture profile.
- Hide obsolete paused-agent warnings on completed or cancelled task
threads.
- Synchronize the company-switch settings test with React effects after
a CI failure. Preserve its security assertions and production remount
behavior.

## Verification

The branch directly includes master `9ef434fe79`. The required `ci /
verify` and `ci / e2e` gates pass on head `765fccd67c`. [Current-head
CI](https://github.com/paperclipai/paperclip/actions/runs/39464348614)
passes typecheck, build, tests, and all eight browser shards. Security
checks pass. [Greptile gives this exact head
5/5](https://github.com/paperclipai/paperclip/pull/16959#issuecomment-6291156549),
with no unresolved review threads. The optional Paperclip Review check
remains queued because its assigned agent is paused. It is not a
required branch check. No check result was overridden.

Local workspace typecheck and build pass after the master merge, along
with database migration checks, token gates, map checks, Storybook
catalog checks, and the Copilot profile identity check. Standard builds
refresh Runner vendor output. Focused checks pass 2,008
API/paging/context tests and 7 strict case protocol tests. The final
commit changes one test fixture only; its ten tests, UI typecheck, and
token gates pass. No production source changed in that commit.

On current head `765fccd67c`, all **194 critical workflow tests**
executed and passed with no skipped tests in those files:

- 38 Stop/ownership tests, including board completion intent, native
retries, stale Stop, and preparation with an old checkout.
- 2 actual legacy process/HTTP handoff tests, including scratch failure,
a closed-stdio child that continues writing, and recovery through a
fresh worker after the process group stops.
- 5 migration tests, including SQL column metadata, preserved null
authority context, and all six prior-numbering receipts.
- 3 process-settlement and 5 scratch tests.
- 88 handoff-service, 18 heartbeat handoff, and 21 native case-tool
integration tests.
- 7 bounded case/document-read projection tests and 7 strict
protocol-contract tests.

The current CI checkout was `fed65d49`, merging `765fccd67c` into master
`9ef434fe79`. Its new document regressions reconstruct exact
Unicode/escaped bodies and revisions through real native authority and
authenticated HTTP with generic API tools disabled. They cover
case-owned, primary-task, and linked-task documents; routing on every
page; stale document snapshots; access loss and unlinking;
unrelated/foreign/private sources; system documents; and quarantined
content. Active-task document tools retain their existing behavior.
These are current-head test results from the first CI attempt; the
browser-only retry did not rerun these suites.

The full `64fdd3403f` CI failed one separate settings test: after a
company switch, it observed an unsettled API-key draft. The production
component and test were unchanged from prior green source. Production
switches remount both keyed settings layers; the test directly reused an
unkeyed view and used a timer tick instead of waiting for React effects.
The final commit replaces that fixture synchronization with React `act`.
The exact secret-clearing assertions and all timeouts remain unchanged.
The original assertion failure remains recorded; no unchanged retry was
used. The narrow unchanged baseline and fixed fixture each pass 10
tests.

The upstream result-paging merge preserves case capture namespaces and
current asset privacy. Its API integration suite executed 39 of 41 tests
successfully, with two skipped; the optional assigned-result PRP
integration test was skipped. Neither skipped suite is counted in the
194 critical tests or as live provider proof.

All 361 upstream migration entries and SQL bodies remain unchanged. Six
case migrations follow master as 0363–0368 with identical SQL hashes.
Official Drizzle generators produced the combined metadata and five
retained snapshots. Generation reports no schema changes, and the
snapshot drift test passes. The earlier `ce361ae5d0` CI passed on
attempt 1, including its then-current 179 critical tests. Fresh review
later identified unusable document references with generic API tools
disabled; the bounded document reads and ten additional authenticated
HTTP tests address that finding.

The full local `pnpm test:run` attempt was cancelled after this Mac
exhausted PostgreSQL semaphore capacity; it is **not** counted as a
pass. Later database attempts also stopped before assertions or skipped,
including one sandbox socket denial. Two original stale-Stop regressions
did reproduce the bug against the old source. A REST serialization
assertion passed after two preserved database-startup failures. No test
timeout was increased. All 194 assertions listed above executed and
passed in current-head CI. The complete 385-file, 3,963-test UI shard
also passed after the fixture repair.

A configured-provider receipt workflow passed 13 backend checks across
triage → implementation → verification → correction → done, using one
task and four work runs. The saved receipt is revision 2 with total 6
USD; case and task both closed. The original canonical attempt remains
failed because its browser assertion confused the terminal destination
with completed-task tooltips. The corrected full browser verifier passed
on the same retained database and original runs, with no additional
provider calls. The last clean-build browser readback is on
`6a86c54a53`; later changes affect recovery, authorization, bounded
reads, and settlement. Accounted work cost is a $1.010574 rate-card
estimate; provider billing and setup-probe usage are unavailable.

Copilot profile 49 passed the unchanged seven-fixture private protocol
pack (46/46 checks, no retries) and the product file-edit canary after
master's completion-replay change. All 60 bound implementation hashes
remain unchanged on the current source. Historical profile-47/48
identities and results remain intact. See the [profile-49 compatibility
record](https://github.com/paperclipai/paperclip/blob/codex/persistent-case-handoffs/doc/plans/2026-10-20-copilot-completion-recovery-compatibility-v49.json).
This is bounded compatibility evidence, not a fresh full matrix,
installed-package product test, or Daytona rollout.

Preserved CI failures and retries:

- On `20662930ac`, Slack managed setup failed a five-second success
assertion after reload. Its test/application path was unchanged from a
prior green head. One unchanged-source retry passed all 25 tests in that
shard; no source or timeout changed.
- On `9402b4f17a`, two CI workers received shutdown signals at 15:24:19
UTC. Neither log had an assertion failure before interruption. One
unchanged-source retry of each passed all Runner static checks and all
32 tests in browser shard 6; both aggregates passed. The next head,
`7b4374edaf`, passed all CI without retries.
- On `295f289b5a`, serialized shard 9 timed out during its first
device-login test, then the next test found an active login lease.
Direct login/test paths were unchanged from a prior green head. Cold
import plus a timed-out async body using the next test's mutable
actor/store is the supported fixture-race explanation; no request trace
proves that sequence. The original failure remains recorded. One
unchanged-source retry passed all 294 tests in 19 files, including all
41 device-login tests. No login source or timeout changed. The
subsequent `ce361ae5d0` CI passed without retries.
- On `64fdd3403f`, one company-switch settings assertion observed an
unsettled draft. The final test-only commit uses React `act` without
changing assertions, timeouts, or production code. On `765fccd67c`, all
ten settings tests and the complete 385-file, 3,963-test UI shard
passed.
- On `765fccd67c`, browser shard 3 completed the split-handoff
assertions, then cleanup received HTTP 500 from run cancellation after a
one-second PostgreSQL row-lock timeout. The extra chat run was reporting
completed work while teardown began. Cleanup stopped before restoring a
global experiment flag, causing a later assertion to fail. Relevant
production and test paths are unchanged from an earlier green commit.
Stop's lock order is consistent, and no nested self-lock was found; the
exact PostgreSQL blocker was not recorded. One unchanged-shard retry
passed. The original failure, trace, and downstream flag failure remain
recorded. No Stop implementation, cleanup assertion, or timeout changed.
The retry supports intermittent cleanup contention; it does not identify
the original blocking transaction.

## Risks

- Local process transfers require host-owned PID/group absence. Missing
or uncertain termination evidence keeps delivery pending, independently
of scratch setup.
- Handoff admission and task cancellation share execution ownership. New
regression tests cover temporary holds, late source writes, and
uncertain provider termination.
- Retained private sources use an intersection of existing permissions.
This intentionally fails closed when a source is missing or an existing
task cannot preserve that boundary.
- Migrations add tables and indexes. They preserve master history and
can replay over prior feature installs. Migration 0368 adds nullable
internal authorization provenance with no default or authority backfill.
Historical transfers still awaiting release without this context fail
closed; an operator must cancel them and request a new handoff.
- Delayed Cloud handoffs depend on the current Cloud membership RPC. An
unavailable, ambiguous, expired, or mismatched response holds the
handoff for retry; it does not reuse cached authority or grant access.
- Bounded reads reload and reauthorize the selected projection per page.
Underlying work grows with the data size and number of pages. Changed
data or routing requires restarting from offset zero without the old
snapshot.
- Phase names and workflow diagrams remain advisory. They do not satisfy
review or approval gates.
- This is a large, cross-cutting PR. The maintainer requested one PR for
the whole feature; the commits separate data, runtime, UI, and
qualification work.

## Model Used

OpenAI Codex, GPT-6 family, with tool use, code execution, and delegated
code review. The exact deployment model ID and context-window size were
not exposed in the session. Live fixture agents used the explicitly
configured provider model recorded in their test evidence.

## Checklist

- [x] I have included a thinking path that traces from project context
to this change
- [x] I have specified the model used (with version and capability
details)
- [x] I have checked ROADMAP.md and confirmed this PR does not duplicate
planned core work
- [x] I have searched GitHub for duplicate or related PRs and linked
them above
- [x] I have either (a) linked existing issues with `Fixes: #` / `Closes
#` / `Refs #` OR (b) described the issue in-PR following the relevant
issue template
- [x] I have not referenced internal/instance-local Paperclip issues or
links (only public GitHub `#NNN` / `github.com/paperclipai/paperclip`
URLs)
- [x] My branch name describes the change (e.g. `docs/...`, `fix/...`)
and contains no internal Paperclip ticket id or instance-derived details
- [x] I have run tests locally and they pass
- [x] I have added or updated tests where applicable
- [x] I have updated relevant documentation to reflect my changes
- [x] I have considered and documented any risks above
- [x] All Paperclip CI gates are green
- [x] Greptile is 5/5 with no open P2s, recommendations, or follow-ups
- [x] I will address all Greptile and reviewer comments before
requesting merge

---------

Co-authored-by: Paperclip <noreply@paperclip.ing>

2026-10-20 12:46:14 -05:00

13 KiB

Raw Blame History

Hiring, agent configuration, and organization — agent access

Product map: Hiring, agent configuration, and organization.

Feature ID: agents.

Review date: 2026-10-12. Source commit: 58e16dbddfc9c8d0cf60867f9d9847f0feda055b. Native REST transport reconciled in this change; see the shared transport contract.

This is a source-reviewed baseline of representative operations and known gaps. Each capability needs operation-level review before its entire scope can be qualified. No live agent execution is claimed. Follow the map contract for format, permission checks, evidence and maintenance.

Scoped runner-default source review: 2026-10-12 at eee0534154eaf5d0d3f2bbc97c1e24d86f34a5a5, limited to creation/configuration, approval activation, built-in provisioning and import preservation described below. The original baseline commit above still governs unrelated entries; this update does not claim operation-complete review or live agent verification.

`hire`

Capability ID: agents/hire.

Review: baseline.

API: present — representative operation IDs:

POST /api/companies/{companyId}/agent-hires.
POST /api/companies/{companyId}/agents.

Agent access: partial — Native hire_agent; REST / CLI agent; runtime inheritance for native hires.

Creation requests can use runner: auto | paperclip | legacy; the server resolves and persists execution configuration. Claude Auto selects the native ACPX profile on qualified macOS and GNU/Linux x64/ARM64 targets; unknown environment qualification remains Legacy. Darwin SSH defaults to Legacy while preserving explicit native assets and saved execution. Saved runner/model/provider overrides survive edits and approval activation. This selection does not grant connection access or bypass hiring authorization. Source: runner resolver.

Policy: Agent creation/configuration uses action grants, resource policies and change consent; some controls require a board actor.

The creation boundary resolves the request-level runner before adapter validation and credential normalization. Omitted/automatic new Codex selection uses the qualified target's default; explicit Legacy is retained. Disabled adapters and external overrides affect new selection, while missing credentials/runtime or incompatible settings remain errors. Existing agents and reviewed pending hires retain their recorded execution. Native hire_agent explicitly inherits its caller's runtime and remains subject to permission and credential restrictions; it is not an arbitrary host-configuration or login tool. REST/CLI creation and hire runner options use the shared request contract.

Evidence: source-only — implementation at the reviewed commit above.

Scoped Evidence: source-only — creation policy, runner resolution, native hire binding, frozen approval activation and same-harness import preservation at the scoped commit. No permitted-agent journey was executed for this map update.

Verification: not-run — exercise a listed permitted operation with actual agent credentials/run context; inspect its result and applicable persistence/attribution. Assert the relevant denied actor, owner, company or mode from Policy. Review the remaining operations before qualifying the whole capability.

For this scope, compare automatic and explicit Legacy creation, assert foreign company/environment and missing-grant denials, then compare pending approval configuration across restart and activation. An unrelated edit/import update must not change existing execution; a new package without a runner choice receives creation defaults. Do not count a board call or component test as agent evidence.

Gaps: Identity/persona hiring works. Operator account login and sensitive adapter/environment configuration are separate guarded actions.

The runner request does not grant new account, environment-probe or native API fallback authority; see validation access and environment administration. Full access qualification remains not run.

`instructions`

Capability ID: agents/instructions.

Review: baseline.

API: present — representative operation IDs:

GET /api/agents/{id}/instructions-bundle/file.
PUT /api/agents/{id}/instructions-bundle/file.

Agent access: available — Native read/update/history/restore instruction tools; REST / CLI.

Policy: Agent creation/configuration uses action grants, resource policies and change consent; some controls require a board actor.

Evidence: source-only — implementation at the reviewed commit above.

Gaps: Canonical entry file, revision guard, company boundary, and protected-change policy apply.

`organization`

Capability ID: agents/organization.

Review: baseline.

API: present — representative operation IDs:

GET /api/companies/{companyId}/org.
PATCH /api/agents/{id}.

Agent access: available — list_agents/get_agent; native call_api for permitted configuration fields; REST / CLI.

Policy: Agent creation/configuration uses action grants, resource policies and change consent; some controls require a board actor.

Evidence: source-only — implementation at the reviewed commit above.

Gaps: Reporting-line changes need configuration authorization; a title alone gives no authority.

`lifecycle`

Capability ID: agents/lifecycle.

Review: baseline.

API: present — representative operation IDs:

POST /api/agents/{id}/pause.
POST /api/agents/{id}/resume.
POST /api/agents/{id}/terminate.
DELETE /api/agents/{id} permanently purges an eligible terminated/rejected agent.
POST /api/agents/{id}/lifecycle/retry.
GET /api/agents/{id} returns the safe failed phase/step and action in lifecycleError.

Agent access: partial — REST / native call_api resume under agent_config:update change-grant authorization; pause and termination remain board-only.

Policy: Agent creation/configuration uses action grants, resource policies and change consent; some controls require a board actor. Permanent deletion is board-only. Retained case handoff references return 409 agent_case_history_retained before dependency cleanup; keep the agent terminated.

Evidence: source-only — implementation at the reviewed commit above.

Gaps: Resume has a generic native REST path. Pause and termination remain enforced actor restrictions.

`built-in`

Capability ID: agents/built-in.

Review: baseline.

API: present — representative operation IDs:

POST /api/companies/{companyId}/built-in-agents/{key}/provision.

Agent access: partial — REST / native fallback for eligible provisioning; built-in routine controls require board.

Policy: Agent creation/configuration uses action grants, resource policies and change consent; some controls require a board actor.

Evidence: source-only — implementation at the reviewed commit above.

Gaps: Provisioning checks agents:create. Enabling/running its built-in routine is not the same permission path.

Scoped source review: built-in provisioning recognizes the underlying Codex harness of a native company agent. New configured provisioning resolves the runner, while unrelated reconciliation and same-harness edits retain existing execution. Verify explicit runner changes preserve compatible saved settings; the general creation/access policy above still applies. This is source-only evidence at the scoped commit, not a live provisioning result.

`primary-agent`

Capability ID: agents/primary-agent.

Review: baseline.

API: present — GET and PUT /api/companies/{companyId}/primary-agent/me.

Agent access: restricted — this is the authenticated human's personal preference. Agent keys and native call_api use agent identity and are denied by the board gate; no dedicated agent selection tool is present.

Policy: Requires board user identity, company access and active membership, except implicit-local board or instance administrator. Viewers may choose their own primary. Selection requires a same-company readable, available agent and is scoped to actor user ID; the request cannot select another user's preference. Explicit null selection is rejected; leave/termination can clear an unavailable choice. Initialization stays recorded so clearing does not silently replace it.

Evidence: source-only — reviewed on 2026-10-13 at 60bc0d5fee9a6e48db08fc016521fee3d9f57d67: routes, service, provider, and route tests.

Verification: not-run — select and read the preference as two company users; assert isolation, wrong-company/unreadable targets, pending/terminated agent rejection and agent-key denial. Verify replacement confirmation and fallback precedence in the UI; inspect the cleared state after permitted leave/termination.

Gaps: No agent mutation path for another person's preference. Live human/agent access is unverified; UI fixture callbacks do not establish persistence or task and Chat default resolution.

`grok-bot`

Capability ID: agents/grok-bot.

Review: baseline.

API: present — POST /api/companies/{companyId}/grok-invitations, operator-owned Grok invitations and agent-scoped /grok-binding setup, routine, verify, stop, recover and disconnect endpoints. Direct HTTPS /mcp/runner uses fixed named-agent OAuth. Grok tools expose durable assignment operations and task-bound upload/download grants.

Agent access: available — paperclip_grok_* through the dedicated agent grant. First-party Paperclip operations execute through the shared Rust Runner. The Bot retains its own apps; Paperclip workspace and third-party app tools are unavailable.

Policy: company/agent/client/resource/authorizer binding; current manage-agent permission for consent and subsequent calls; normal task ownership, controller lease, governance, pause, revocation and recorded budget checks. Execution and file transfers require current task-bound sessions. Secret credentials remain in encrypted secret storage and are redacted from HTTP logs.

Evidence: source-only — routes, shared authority, routine receipts and Runner broker. Temporary actual-client OAuth and an outbound routine wake are qualified separately; they do not prove this candidate’s acceptance.

Verification: not-run — deployed acceptance pending; focused protocol/authorization and actual Rust backend tests; run fresh/saved setup, assigned and proactive tasks, duplicate calls, stale sessions, restart, ambiguous delivery, stop and recovery through the real client on the serving managed candidate.

Gaps: deployed two-way work and unattended first-time onboarding remain unqualified; remote termination, provider transcript streaming and usage accounting are unavailable.

13 KiB Raw Blame History Unescape Escape

Hiring, agent configuration, and organization — agent access

hire

instructions

organization

lifecycle

built-in

primary-agent

grok-bot

13 KiB

Raw Blame History

`hire`

`instructions`

`organization`

`lifecycle`

`built-in`

`primary-agent`

`grok-bot`